Backing up one more step...

[William S. Frantz]

> To see the problem, think through what happens if either district
> fails after the message is sent.  In particular, you must ensure that
> both districts involved in the message agree about whether the message
> has been sent in the event of rollback.  If you start thinking this
> through, you'll discover that rollbacks cascade in remarkably ugly
> ways unless you introduce message replay logic.

KeyTXF had message replay logic.  Generalized message replay logic
introduces its own can of worms however.  Sometimes, like SIGINT, 
you want the replay to interrupt the program at the same place it
was interrupted on play.  KeyTXF only replayed non-time (just order)
dependent events.

-----------------------------------------------------------------
Bill Frantz                   Periwinkle  --  Computer Consulting
(408)356-8506                 16345 Englewood Ave.
frantz@netcom.com             Los Gatos, CA 95032, USA